Analysis of Agricultural Commodities Prices with New Bayesian Model Combination Schemes

Abstract

In the described research three agricultural commodities (i.e., wheat, corn and soybean) spot prices were analyzed. In particular, one-month ahead forecasts were built with techniques like dynamic model averaging (DMA), the median probability model and Bayesian model averaging. The common features of these methods are time-varying parameters approach toward estimation of regression coefficients and dealing with model uncertainty. In other words, starting with multiple potentially important explanatory variables, various component linear regression models can be constructed. Then, from these models an averaged forecast can be constructed. Moreover, the mentioned techniques can be easily modified from model averaging into a model selection approach. Considering as benchmark models, time-varying parameters regression with all considered potential price drivers, historical average, ARIMA (Auto-Regressive Integrated Moving Average) and the naïve forecast models, the Diebold–Mariano test suggested that DMA is an interesting alternative model, if forecast accuracy is the aim. Secondly, the interpretation of time-varying weights ascribed to component models containing a given variable suggested that economic development of emerging BRIC economies (Brazil, Russia, India and China) is recently one of the most important drivers of agricultural commodities prices. The analysis was made on the monthly data between 1976 and 2016. The initial price drivers were various fundamental, macroeconomic and financial factors.

Agricultural commodity price dynamics and their determinants: A comprehensive econometric approach

“…categories (i) market fundamentals, (ii) macroeconomic factors, (iii) financial variables and (iv) determinants related to climatic variability. The covariates correspond to predictors used in the literature aimed at modelling and forecasting agricultural commodity prices (see, e.g., the battery of variables employed in Drachal, 2019). For the category of market fundamentals, they include total production figures at the global level and for the most important producers for each one of the grains, yields and stock-to-use ratio.…”

“…Ahumada and Cornejo (2016) show that the strong correlation observed in the prices of corn, soybeans and wheat can be utilized to significantly improve forecasting accuracy in time series specifications. In addition to univariate and multivariate time series models, the literature on commodity price forecasting has also employed artificial neural networks (Kohzadi et al, 1996), models aimed at modelling the dynamics of the second moment of the commodity price time series (Bernard et al, 2008) for prediction and model averaging schemes (Drachal, 2019). In particular, models that account for the particular dynamics of the volatility of agricultural commodity prices appear to robustly improve probabilistic forecasts of price changes, as shown in Ramirez and Fadiga (2003).…”

“…Our results concerning the response of agricultural commodity prices to their determinants highlight the role played by real exchange rate dynamics and market fundamentals, as well as by global macroeconomic and financial shocks as captured by US industrial production and stock market fluctuations. We find scattered 1 The analysis in Drachal (2019) is also applied to the prices of these particular agricultural commodities. evidence for differences in out-of-sample predictive performance across states of the business cycle.…”

Prediction of Soybean Price Trend via a Synthesis Method With Multistage Model

“…Zhang et al (2018) applied a quantile regression-radial basis function (QR-RBF) neural network model to predict of soybean price in China. Drachal (2019) proposed a new Bayesian model combination schemes for analysis of soybean price. These researches mentioned above are about to forecast univariate time series.…”

Price Forecasting and Span Commercialization Opportunities for Mexican Agricultural Products

“…Moreover, the usefulness of these approaches depends of the local market context. For example, in China, the high volatility of the prices lead to ANN models or hybrid models to get a better fit than the Box-Jenkins methodology [7,8].…”